Impact Evaluation of University Grants

The University of Florence (Italy) offers some scholarships every year to eligible freshmen: eligibility is based on merit and economic needs. The number of grants awarded is usually constrained by a fixed budget: not all students applying for a grant receive one. Applicants are ranked on the basis of an economic indicator, that depends in a deterministic way on family income, property and personal assets, and family structure, and only a varying percentage of high-ranking applicants receive a grant. The amount of aid received varies with the value of their economic indicator. We investigate whether the grant is an effective tool to prevent students from low-income families from dropping out of higher education. We show which causal parameters can be identified from available information: under mild assumptions, the grant’s assignment rule can be exploited to estimate the grant’s effects at different values of the economic indicators. Preliminary results for two cohorts of freshmen seem to suggest that the grants have significant effects, reducing the dropout probability, although some grant's amounts are too low to prevent low-income students from dropping out.